System Thinking · Personal Growth
Why improvement initiatives stall, and how to make change stick

Walk into almost any mid-sized company and ask for the list of improvement initiatives currently running. You will usually get a long one: a lean project in the plant, a cost-reduction drive, a new ERP module, a sales-force effectiveness programme, a digital project or two.
Then ask a second question: which of these have changed the way people work on the ground, and shown up in the profit and loss account?
The list gets very short.
Initiatives that live only on paper
I have observed a familiar life cycle for most improvement initiatives:
- A problem is noticed and a solution is chosen, often from a benchmark or a consultant’s study.
- The solution is designed by a small team, usually away from the people who will have to run it.
- Standard operating procedures are written and announced.
- A few months later, the reports still say the initiative is running. On the floor, people are working much as before.
- Management attention moves on. The initiative fades, or is quietly put “on hold”.
Nobody declares failure. It just stops mattering.
Why does this happen so often? It is not because people don’t know what the target state should be. In my experience, the people closest to the work usually understand the problem better than anyone else. So if they know what should change, why doesn’t it change?
The missing “how”
The answer usually lies in two gaps.
Gap 1: the “how” is never answered at the individual level. A goal like “reduce lead time by half” is clear at the company level. But what exactly should the planner do differently on Monday morning? What should the supervisor stop doing? What should the salesperson promise a customer? Unless each person knows what changes in their own way of working, nothing changes.
Gap 2: the people who must change had no say in the design. When a solution is designed behind closed doors and handed down as rules, two things happen:
- Side effects go unnoticed. The new rule may solve a problem in one area and create a bigger one elsewhere, because the people who could have spotted it weren’t asked.
- There is no ownership. People feel the change was forced on them. When the first difficulty comes up (and it always does), nobody feels responsible for solving it. The easiest path is to go back to the old way.
Resistance is not stubbornness
It is easy to label this as “resistance to change”. But Dr. Eliyahu M. Goldratt offered a more useful way to look at it. He described resistance as a series of layers, each a legitimate doubt that must be answered before people will commit:
- We don’t agree on what the problem is.
- We don’t agree that this is the right direction for the solution.
- We aren’t convinced the solution will actually deliver the results.
- We can see it causing problems elsewhere.
- We can see obstacles in implementing it.
- We have doubts we haven’t voiced yet.
Seen this way, “resistance” is not something to be overcome by pressure. It is information. Each objection points to something the solution has not yet answered.
A better way to make change happen
Involve the people who will run it
Bring together the people whose work will change, or representatives from every affected area, as part of the design team. Don’t hand them a finished solution. Give them a clear direction and let them work out the details.
Debate the solution openly
Before any procedure is written, put the direction through tough questions:
- What problem does this actually solve?
- Why this solution and not another?
- What could it break in another department?
- What exactly will each person do differently?
- What happens when conditions on the ground change?
This debate takes time at the start. Managers often find it frustrating. But it is where people move from “their idea” to “our idea”.
Answer the “how” down to each role
Translate the direction into changes in measures, rules and daily behaviour for each level: top management, department heads, supervisors and the front line. If you can’t say what a role will do differently, the design is not finished.
Here is what that can look like for a single goal, halve the order lead time, in a plant:
| Role | Stops doing | Starts doing |
|---|---|---|
| Planner | Releasing every order as soon as it is received | Releasing an order only a fixed number of days before its due date |
| Supervisor | Picking the job with the easiest set-up | Running jobs in the order of their buffer colour, red first |
| Salesperson | Promising the date the customer asks for | Quoting dates from the load on the constraint |
| Plant head | Reviewing each machine’s efficiency | Reviewing on-time delivery and the top reasons orders turned red |
A table like this is not a formality. Each line is a change in someone’s daily habit, and each one needs that person’s agreement. Making the table together is half the change.
Expect a slower start and a faster finish
A change designed this way takes longer to get going. But once agreed, it moves very quickly, because the people implementing it designed it and want it to work. When problems come up, they solve them instead of abandoning the change.
Change imposed is temporary. Change owned sustains itself.
In short
Before launching your next initiative, ask yourself:
- Do the people who must change agree on the problem?
- Did they help design the solution?
- Can each person say what they will do differently?
- Have we looked for the side effects in other areas?
If the honest answer to any of these is no, you may be about to add one more initiative to the list that lives only on paper.